Determining a way to adequately sample sparse populations remains a challenge in ballast water testing. In an effort to quantify variability in counting, the Naval Research Laboratory conducted an experiment which eliminated as much variability as possible by use of standardized micro beads in artificial seawater. Concentrations counted were based on Environmental Technology Verification Program, International Maritime Organization and additional proposed guidelines, all of which involved sparse populations. Inert 10-micron beads at 1, 5, 10, 50, 100, 500 and 1000 beads per milliliter (mL) were used to represent phytoplankton while 100-micron beads at 10, 30 and 60 beads per 500 mLs of water represented zooplankton. Combinations of initial sample collected and level of concentration were tested as well as observer bias. The results showed that for phytoplankton beads, the 50, 100 and 1000 concentrations were not significantly different and had accuracy and precision measurements below the 10% and 20% benchmarks respectively. The zooplankton populations all showed poor precision with all measurements well above the 20% benchmark. In both experiments, no significant differences were found between observer counts.